DESCRIPTION (provided by applicant: It is proposed to investigate the use of a computer-controlled, autonomous, robot to perform a specified surgical procedure-mastoidectomy. Mastoidectomy, a core surgical procedure in Otolaryngology-Head and Neck Surgery, is chosen because of its high frequency (conservative estimate of at least 100,000 procedures performed annually in the United States), the unique anatomy involved (vital structure encased in bone which does not deform during surgical intervention), the surgical technique (performed using a drill and particularly amenable to robotic application), and the principal investigator's expertise in image-guided otologic surgery. In addition to otologic surgery, however, the proposed research has widespread applications to other surgical fields that incorporate robotic intervention. [unreadable] [unreadable] To accomplish the overall goal of a robotic mastoidectomy, a graduated approach is planned as follows: (1) Analysis of human operator strategy in performing mastoidectomies. Before programming a robot to perform a human task, it is necessary to understand how a human typically performs that task. To that end, the motion of a drill used by surgeons to mill away bone as they perform mastoidectomies in the temporal bone laboratory will be tracked via an image-guided surgical system. The time course of the recorded motion will be analyzed via decomposition into action primitives; (2) Robotic mastoidectomy milling based on optimized human motions. The information from Specific Aim 1 will be utilized to develop optimized milling strategies by minimizing time and distance of travel. Robotic milling with optimized strategies will be compared to human milling; (3) Exploration of a new drilling strategy enabled by image-guided surgery (IGS) - en bloc removal of tissue. Because IGS provides "x-ray vision," a unique strategy will be examined, namely, fitting an optimally- shaped polyhedron to the mastoid volume followed by edge routing with subsequent en-bloc removal of the tissue. It is hypothesized that this strategy will be more efficient than either the human or robotic interventions examined in Specific Aim 2. [unreadable] [unreadable] [unreadable]